Author: cade metz.cade metz business date of publication: 11.10.15.11.10.15 time of publication: 7:00 am.7:00 am from wired.com
IN OPEN SOURCING its
artificial intelligence engine—freely sharing one of its most important
creations with the rest of the Internet—Google showed how
the world of computer software is changing. These days, the big Internet giants
frequently share the software sitting at
the heart of their online operations. Open source accelerates the
progress of technology. In open sourcing its TensorFlow
AI engine, Google can feed all sorts of machine-learning research
outside the company, and in many ways, this research will feed back into
Google.
But Google’s AI engine also
reflects how the world of computer hardware is
changing. Inside Google, when tackling tasks like image recognition andspeech recognition and language translation,
TensorFlow depends onmachines equipped with GPUs, or graphics processing
units, chips that were originally designed to render graphics for games and the
like, but have also proven adept at other tasks. And it depends on these chips
more than the larger tech universe realizes.
According to Google engineer Jeff Dean, who helps oversee the company’s AI work,
Google uses GPUs not only in training its artificial intelligence services, but
also in running these services—in delivering them to the
smartphones held in the hands of consumers.
That represents a significant shift.
Today, inside its massive computer data centers, Facebook uses GPUs to train
its face recognition services, but when delivering these services to
Facebookers—actually identifying faces on its social networks—it uses
traditional computer processors, or CPUs. And this basic setup is the industry
norm, as Facebook CTOMike “Schrep” Schroepfer recently
pointed out during a briefing with reporters at the company’s Menlo Park,
California headquarters. But as Google seeks an ever greater level of
efficiency, there are cases where the company both trains and executes its AI models on GPUs inside the data
center. And it’s not the only one moving in this direction. Chinese search
giant Baidu is building a new AI system that works in much the same way. “This
is quite a big paradigm change,” says Baidu chief scientist Andrew Ng.
The change is good news for nVidia, the chip giant that specialized in GPUs. And it
points to a gaping hole in the products offered by Intel, the world’s largest
chip maker. Intel doesn’t build GPUs. Some Internet companies and
researchers, however, are now exploring FPGAs, or field-programmable
gate arrays, as a replacement for GPUs in the AI arena, and Intel recently acquired a
company that specializes in these programmable chips.
The bottom line is that AI is playing an increasingly
important role in the world’s online services—and alternative chip
architectures are playing an increasingly important role in AI. Today, this is
true inside the computer data centers that drive our online services, and in
the years to come, the same phenomenon may trickle down to the mobile devices
where we actually use these services.
Deep Learning in Action
At places like Google, Facebook, Microsoft, and Baidu, GPUs have
proven remarkably important to so-called “deep learning” because they can
process lots of little bits of data in parallel. Deep learning relies on neural networks—systems
that approximate the web of neurons in the human brain—and these networks are
designed to analyze massive amounts of data at speed. In order to teach these
networks how to recognize a cat, for instance, you feed them countless photos
of cats. GPUs are good at this kind of thing. Plus, they don’t consume as much
power as CPUs.
But, typically, when these companies put deep learning into
action—when they offer a smartphone app that recognizes cats, say—this app is
driven by a data center system that runs on CPUs. According to Bryan Catanzaro,
who oversees high-performance computing systems in the AI group at Baidu,
that’s because GPUs are only efficient if you’re constantly feeding them data,
and the data center server software that typically drives smartphone apps
doesn’t feed data to chips in this way. Typically, as requests arrive from
smartphone apps, servers deal with them one at a time. As Catanzaro explains,
if you use GPUs to separately process each request as it comes into the data
center, “it’s hard to get enough work into the GPU to keep it running
efficiently. The GPU never really gets going.”
MORE ARTIFICIAL INTELLIGENCE
That
said, if you can consistently feed data into your GPUs during this execution
stage, they can provide even greater efficiency than CPUs. Baidu is working
towards this with its new AI platform. Basically, as requests stream into the
data center, it packages multiple requests into a larger whole that can then be
fed into the GPU. “We assemble these requests so that, instead of asking the
processor to do one request at a time, we have it do multiple requests at a
time,” Catanzaro says. “This basically keeps the GPU busier.”
It’s unclear how Google approaches this issue. But the
company says there are already cases where TensorFlow runs on GPUs during the
execution stage. “We sometimes use GPUs for both training and recognition,
depending on the problem,” confirms company spokesperson Jason Freidenfelds.
That may seem like a small thing. But it’s actually a big
deal. The systems that drive these AI applications span tens, hundreds, even
thousands of machines. And these systems are playing an increasingly large role
in our everyday lives. Google now uses deep learning not only to identify
photos, recognize spoken words, and translate from one language to another, but
also to boost search results. And other companies are pushing the same
technology into ad targeting, computer security, and even applications that
understand natural language. In other words, companies like Google and Baidu
are gonna need an awful lot of GPUs.
AI Everywhere
At the same time, TensorFlow is also pushing some of this AI
out of the data center entirely and onto the smartphones themselves.
Typically, when you use a deep learning app on your phone,
it can’t run without sending information back to the data center. All the AI
happens there. When you bark a command into your Android phone, for instance,
it must send your command to a Google data center, where it can processed on
one of those enormous networks of CPUs or GPUs.
But Google has also honed its AI engine so that it, in some
cases, it can execute on the phone itself. “You can take a model description
and run it on a mobile phone,” Dean says, “and you don’t have to make any real
changes to the model description or any of the code.”
This is how the company built its Google Translate app.
Google trains the app to recognize words and translate them into another
language inside its data centers, but once it’s trained, the app can run on its
own—without an Internet connection. You can point your phone a French road
sign, and it will instantly translate it into English.
That’s hard to do. After all, a
phone offers limited amounts of processing power. But as time goes on, more and
more of these tasks will move onto the phone itself. Deep learning software
will improve, and mobile hardware will improve as well. “The future of deep
learning is on small, mobile, edge devices,” says Chris Nicholson, the founder
of a deep learning startup called
Skymind.
GPUs, for instance, are already
starting to find their way onto phones, and hardware makers are always pushing
to improve the speed and efficiency of CPUs. Meanwhile, IBM is building a
“neuromorphic” chip that’s designed specifically for AI tasks, and
according to those who have used it, it’s well suited to mobile devices.
Today, Google’s AI engine runs on server CPUs and GPUs as
well as chips commonly found in smartphones. But according to Google engineer
Rajat Monga, the company built TensorFlow in a way that engineers can readily
port it to other hardware platforms. Now that the tool is open source,
outsiders can begin to do so, too. As Dean describes TensorFlow: “It should be
portable to a wide variety of extra hardware.”
So, yes, the world of hardware is
changing—almost as quickly as the world of software=================================================== Have you ever wondered how it would be if your email suddenly came to life? You are about to find out.
====================================================
===========================================================
**Important note** - contact our sister company for very powerful solutions for IP management (IPv4 and IPv6, security, firewall and APT solutions:
www.tabularosa.net
In addition to this blog, Netiquette IQ has a website with great assets which are being added to on a regular basis. I have authored the premiere book on Netiquette, “Netiquette IQ - A Comprehensive Guide to Improve, Enhance and Add Power to Your Email". My new book, “You’re Hired! Super Charge Your Email Skills in 60 Minutes. . . And Get That Job!” will be published soon follow by a trilogy of books on Netiquette for young people. You can view my profile, reviews of the book and content excerpts at:
www.amazon.com/author/paulbabicki
If you would like to listen to experts in all aspects of Netiquette and communication, try my radio show on BlogtalkRadio Additionally, I provide content for an online newsletter via paper.li. I have also established Netiquette discussion groups with Linkedin and Yahoo. I am also a member of the International Business Etiquette and Protocol Group and Minding Manners among others. Further, I regularly consult for the Gerson Lehrman Group, a worldwide network of subject matter experts and have been a contributor to numerous blogs and publications.
Lastly, I
am the founder and president of Tabula
Rosa Systems, a company that provides “best of breed” products for network,
security and system management and services. Tabula Rosa has a new blog and Twitter site which offers great IT
product information for virtually anyone.
==============================================
No comments:
Post a Comment